The realization of the α-trimmed mean filter on the threshold decomposition architecture is described. The threshold decomposition architecture is developed for facilitating realization of nonlinear, order-statistic filters. In this architecture, an input sequence is thresholded at different levels to form a number of binary sequences. The filter output is formed by combining the decision results made at various levels; at each level it is determined whether the signal should be above the associated threshold level. The decision rule used is the distinguishing feature of an individual filter; the current implementation of the α-trimmed mean filter is based on using a threshold logic function as the decision rule, thus allowing the decision rule to output multiple values. The statistical properties and performance of the a-trimmed mean filter are analyzed on the basis of this realization.